PDA

View Full Version : The real problem with the offense



Ltlabner
07-18-2008, 07:14 AM
Hint: It's not Dunn's strike outs.

Thus far the Reds have had 707 Plate Apperances by players who are below replacement level. All 6 of these players have impressivly acomplished the feat of having all negative numbers in VORP, RAR (Runs Above Average) and RAP (Runs Above Position). IOW, Compared to a replacement player, an average player and average players at their position.....these guys are worse.

Nearly 20% of the total PA's of the entire team (exluding pitchers) are from this stellar group of batsmen.

I know CPatt is a favorte target but I just have to say it. He'd have to catch every ball hit past the infield for the Reds, Moeller High School and the local softball beer-league *at the same time* for his defense to ever outweigh his bat.



# YEAR NAME PA EqA OBP SLG VORP RAR RAP

7. 2008 Paul Bako 228 .226 .296 .355 -3.8 -0.8 -5.8
9. 2008 Corey Patterson 212 .197 .225 .337 -10.6 -6.9 -16
13 2008 Javier Valentin 76 .207 .276 .314 -2.3 -1.6 -5.0
14 2008 Paul Janish 72 .162 .278 .203 -4.1 -3.8 -5.4
15 2008 Scott Hatteberg 61 .173 .262 .231 -5.2 -3 -6.4
16.2008 Norris Hopper 58 .181 .286 .200 -3.7 -2.4 -5.1


Next you have another 694 PA's by guys who have been decidely average at the plate. So that's another 31% of the PA's that are just about neutral compared to your run-of-the mill replacement guy from AAA who plays one year and is gone.

Bruce I'm willing to let slide for obvious reasons. Ross is heating up but is the oppisate of a long term solution. Kepp is filling a hole this year, but please don't go all Ryan Freel with the contracts Walt. Jr's a first ballot HOF, but the party is over unfortunatley.




# YEAR NAME PA EqA OBP SLG VORP RAR RAP

2. 2008 Ken Griffey Jr. 374 .267 .348 .400 3.4 14 -2.3
6. 2008 Jeff Keppinger 238 .265 .353 .388 9.3 8.4 2.8
10.2008 Jay Bruce 182 .265 .335 .429 3.7 6.7 -0.6
11 2008 Ryan Freel 143 .250 .340 .359 1.8 3 -2.6
11 2008 David Ross 131 .287 .389 .417 6.2 7.9 4.8


Then you have the guys who are actually earning their paycheck at the plate. Not much to say here other than Hairston is a mirage and we don't know what will happen with Dunner. Dunn could raise his BA and reduce his K's and it still wouldn't do anything to adress the other 50% of PA's that are worthless.



# YEAR NAME PA EqA OBP SLG VORP RAR RAP

1. 2008 Brandon Phillips 403 .280 .323 .476 20.5 22.1 8.6
3. 2008 Adam Dunn 368 .313 .380 .538 24.0 33.5 18.8
4. 2008 Joey Votto 343 .280 .350 .464 13.1 18.5 1.3
5. 2008 E Encarnacion 337 .284 .341 .481 13.5 19.7 5.5
8. 2008 Jerry Hairston 213 .315 .398 .495 24.2 20.2 14


To recap:
20% of plate apperances are worse than replacement/average
31% of the PA's are average
48% of the PA's are above average

(note, it doesn't add up to 100% because pitchers have been excluded).

Conclusion: With half your plate apperances being average or worse you will be sitting at home in October irrespective of what happens with the big Texan.

PuffyPig
07-18-2008, 07:35 AM
I would have thought that an average offense would have approximately 1/3rd of it's players above average, one third average, and one third below average. And our offense is about average.

That's the nature of averages.

We have half our team above average, and only 20% below average. Why isn't our offense above average?

Spring~Fields
07-18-2008, 07:49 AM
Thus far the Reds have had 707 Plate Apperances by players who are below replacement level. All 6 of these players have impressivly acomplished the feat of having all negative numbers in VORP, RAR (Runs Above Average) and RAP (Runs Above Position). IOW, Compared to a replacement player, an average player and average players at their position.....these guys are worse.

Nearly 20% of the total PA's of the entire team (exluding pitchers) are from this stellar group of batsmen.

I know CPatt is a favorte target but I just have to say it. He'd have to catch every ball hit past the infield for the Reds, Moeller High School and the local softball beer-league *at the same time* for his defense to ever outweigh his bat.


Were there other alternatives (players) to be used ?

boognish
07-18-2008, 08:01 AM
Good to see you posting, Ltlabner!

Looking strictly at the rates posted in your analysis, the only solutions in house (defined conveniently by me as giving the majority of the PA which were below replacement level to players in the average category) are replacing the Bako/Patterson PA with Ross and Bruce, respectively. The team's OPS+ before last night--adjusted for park effects--sat squarely at a below-average 95.

There seems to be more intuitive faith that Bruce will improve upon his numbers than can Ross, but I posit that Ross can approximate a league-average productivity level with a goodly number of walks supplemented by the occasional home run along the lines of his second half of 2006 (.203/.325/.484) with a roughly .225/.330/.430 triple slash line.

Is there a level of faith that Encarnacion and Keppinger can both improve upon their first-half production? It is no accident that the three players for whom I posited this question all hit right-handed ;)

edabbs44
07-18-2008, 08:06 AM
I would have thought that an average offense would have approximately 1/3rd of it's players above average, one third average, and one third below average. And our offense is about average.

That's the nature of averages.

We have half our team above average, and only 20% below average. Why isn't our offense above average?

Maybe two things:

1) Situational hitting
2) The level of how under average and over average some of those guys are. Corey Patterson and Jerry Hairston would be one below and one above avg player, yet their compilation would easily be below average.

boognish
07-18-2008, 08:16 AM
Maybe two things:

1) Situational hitting
2) The level of how under average and over average some of those guys are. Corey Patterson and Jerry Hairston would be one below and one above avg player, yet their compilation would easily be below average.

Spot on. Distribution of production, i.e. which situations, and "depth of below replacement crater" or "Patterson effect" seem to be the two places to start.

The question becomes how to fix the problem? I always like to replace the worst players with average players, but the Reds are in a position for a shift in the team constuction at midseason and this offseason, and there is a huge question mark regarding one of the above average performers' location of employment next season.

Ltlabner
07-18-2008, 08:31 AM
We have half our team above average, and only 20% below average. Why isn't our offense above average?

Lots of deadweight near the top of the lineup while the real producers are near the bottom. It also doesn't account for what the other team is doing or how well they might be constructed.

The real producers on the team tend to be feast or famine guys.

This only accounts for our offensive production, you have to factor in defense and pitching to get to wins.

The randomness of baseball games.

Spring~Fields
07-18-2008, 09:15 AM
Lots of deadweight near the top of the lineup while the real producers are near the bottom.


So were the wrong players getting most of the PA, "deadweight near the top" ?

Should the "real producers" have been closer to the top, and the "deadweight" near the bottom?

Would it have made any difference?

Cyclone792
07-18-2008, 10:10 AM
Hint: It's not Dunn's strike outs.

Thus far the Reds have had 707 Plate Apperances by players who are below replacement level. All 6 of these players have impressivly acomplished the feat of having all negative numbers in VORP, RAR (Runs Above Average) and RAP (Runs Above Position). IOW, Compared to a replacement player, an average player and average players at their position.....these guys are worse.

Nearly 20% of the total PA's of the entire team (exluding pitchers) are from this stellar group of batsmen.

I know CPatt is a favorte target but I just have to say it. He'd have to catch every ball hit past the infield for the Reds, Moeller High School and the local softball beer-league *at the same time* for his defense to ever outweigh his bat.


# YEAR NAME PA EqA OBP SLG VORP RAR RAP

1. 2008 Brandon Phillips 403 .280 .323 .476 20.5 22.1 8.6
3. 2008 Adam Dunn 368 .313 .380 .538 24.0 33.5 18.8
4. 2008 Joey Votto 343 .280 .350 .464 13.1 18.5 1.3
5. 2008 E Encarnacion 337 .284 .341 .481 13.5 19.7 5.5
8. 2008 Jerry Hairston 213 .315 .398 .495 24.2 20.2 14



Good stuff.

This is precisely the point most fans lose amongst the trees when complaining about a team. Far too often the blame is placed upon the team's star/established players when in reality the blame should be heaped upon the team's inability to weed out bad players.

Add up the negative VORP with that group of players: it's roughly 30 runs. That's three wins by itself. Even if those players were just replacement level, that would turn the Reds below average offense into an average offense. Heck, if those players were slightly above replacement level, the Reds would be sporting an above average offense right now.

The quickest (and cheapest) solution to improving a team is upgrading the team's lousy players. And the Reds have an abundance of lousy players who need to be dumped into the Ohio and upgraded ... even if the upgrade is barely above replacement level.

Ltlabner
07-18-2008, 11:54 AM
I've dug a little deaper...

I compared each player with at least 100AB's to the other NL players by position. You can see where each player ranked relative to other average players at that position in the same number of outs. (Runs Above Position)

For example: of all the second-basemen in the NL, Brandon Phillips ranks 3rd for scoring runs above and beyond what an average 2nd basemen would produce in the same number of outs.

I listed the players by where they ranked in the NL in decending order.




Player League Rank
by RAP

BPhil 3rd
Dunn 4th
Harriston 5th (as SS)
Edwin 6th
Ross 7th
Kepp 10th
Votto 11th
Bruce 12th (as RF)
Jr 13th
Bako 21st
Patterson 21st

So if you total the number of plate apperances by guys ranked 10th or lower compared to their fellow NL players you get....1365 PAs. That's 40% of the total PA's thus far being significantly less productive compared to other NL teams, by position. That's a bit misleading because Votto is having a nice (not great, but nice) year but he's got Howard, Berkman and Sir Albert to contend with. But you get the idea. Wouldn't be a big deal if Votto was "just" an above average 1B at the plate if you didn't have to contend with 212 sucktastic PA's by CPatt or 228 by Bako.

It might help explain that while about 50% of our total teams PA's have been by generally productive hitters that other teams have (1) more productive, productive hitters and (2) fewer or not as horrable non-productive hitters.

*BaseClogger*
07-18-2008, 01:19 PM
I would have thought that an average offense would have approximately 1/3rd of it's players above average, one third average, and one third below average. And our offense is about average.

That's the nature of averages.

We have half our team above average, and only 20% below average. Why isn't our offense above average?

That's not how averages work. Averages are not the middle of the data... that's a median. Unlike medians, averages are effected by influential points like outliers that can move it away from the center. Half of the data does not have to be on each side of an average.

I'm guessing the below average guys are much further away from true average production than our above average guys...

Spring~Fields
07-18-2008, 02:21 PM
That's not how averages work. Averages are not the middle of the data... that's a median. Unlike medians, averages are effected by influential points like outliers that can move it away from the center. Half of the data does not have to be on each side of an average.

Would that be a good explanation why ERA, BA or RBI alone would tend to deceive me if I used those alone to make determinations about a pitcher or batter? Or even more extreme use averages to compare the team against itself without bringing in the rest of the league?

I have a lot to learn, always will, that's just life. Question: When we look at a pitcher, or batter, we see ERA or BA, that kind of tosses everything in accumulatively, which can be misleading, do they have a quick stat lookup that is ERA without outliers? Or BA without outliers? I am guessing OBP is an adjustment for that ?

RedsManRick
07-18-2008, 03:36 PM
Question: When we look at a pitcher, or batter, we see ERA ?

For BA, there are so many plate appearances that no one plate appearance can be considered an outlier.

For any league wide stat, there are usually so many players that outliers aren't terribly influential.

However, for ERA, since it's based on so few appearances, one or two bad outings can kill your ERA. As far as starters go, this is one thing that QS helps alleviate.

One Arroyo-esque outing can hide the fact that 2/3 of the time he's been pretty effective. And since SP affect individual games, that matters.

Spring~Fields
07-18-2008, 04:32 PM
For BA, there are so many plate appearances that no one plate appearance can be considered an outlier.

For any league wide stat, there are usually so many players that outliers aren't terribly influential.

However, for ERA, since it's based on so few appearances, one or two bad outings can kill your ERA. As far as starters go, this is one thing that QS helps alleviate.

One Arroyo-esque outing can hide the fact that 2/3 of the time he's been pretty effective. And since SP affect individual games, that matters.

Since some of those writers articles cued or triggered me to look closer at Harang for example, linking that to your thoughts or example of Arroyo, I thought a glance of Harang, his ERA, 4.76 and a 3-11 win-loss was very misrepresentative of his truer performance this year to date where I believe that he had 13 outstanding starts within the set of 20, 13, which could not be just tossed away, like the writers seem to do when attempting to pose their views for the public consumption (13 leaving 7 where maybe 3-5 of those could even be debated ). This ERA, 4.76 and a 3-11 win-loss is completely deceiving me if I don't look deeper.

On BA I guess I was thinking of those feast nights, where player x gets a boost in BA from a 3-4 or 4-5 night against a struggling pitcher that might not be there on most times to accommodate the hitters BA. Or the cases of pitchers, pitching around the better or greater threat batters in a league.

Anyway I was thinking that his, *BaseClogger*'s comment on average was pretty good to show why using the average can be a problem. So to me that would be a good explanation why methods or measurement had to be improved upon, leading to the development of the tools that some of you often use and as Ltlabner or Cyclone is above.

RedsManRick
07-18-2008, 06:07 PM
Since some of those writers articles cued or triggered me to look closer at Harang for example, linking that to your thoughts or example of Arroyo, I thought a glance of Harang, his ERA, 4.76 and a 3-11 win-loss was very misrepresentative of his truer performance this year to date where I believe that he had 13 outstanding starts within the set of 20, 13, which could not be just tossed away, like the writers seem to do when attempting to pose their views for the public consumption (13 leaving 7 where maybe 3-5 of those could even be debated ). This ERA, 4.76 and a 3-11 win-loss is completely deceiving me if I don't look deeper.

On BA I guess I was thinking of those feast nights, where player x gets a boost in BA from a 3-4 or 4-5 night against a struggling pitcher that might not be there on most times to accommodate the hitters BA. Or the cases of pitchers, pitching around the better or greater threat batters in a league.

Anyway I was thinking that his, *BaseClogger*'s comment on average was pretty good to show why using the average can be a problem. So to me that would be a good explanation why methods or measurement had to be improved upon, leading to the development of the tools that some of you often use and as Ltlabner or Cyclone is above.

I did some work on this last year showing the distribution of starts by adjusted ERA for each of our starters. I had a set of excel charts and everything. I'll see if I can dig it up.

The basic point was that Harang's greatest strength was in limiting the number of times he had horrible starts. He was actually less dominant that our other starters, but he gave us the chance to win more often than any body else.

Degenerate39
07-19-2008, 01:23 AM
I hope I'm not the only one that has no clue about what those stats are or mean or anything for that matter.

Jpup
07-19-2008, 02:38 AM
I hope I'm not the only one that has no clue about what those stats are or mean or anything for that matter.

I like stats, but I think it tends to get out of hand on occasion as well.

RedlegJake
07-19-2008, 04:21 AM
I hope I'm not the only one that has no clue about what those stats are or mean or anything for that matter.


I like stats, but I think it tends to get out of hand on occasion as well.

Stats are often done to ridiculous extremes if you're just looking for the big picture. Take every stat in this thread and boil them down to a conclusion. The problem isn't good players - it's too many very poor players receiving too many at bats/innings. Some of those players are simply young, ie Votto, Bruce and Cueto, Thompson and Bailey. They'll improve. It's the CPatts, Foggs, Bako, Janish types that don't project to ever be better that the Reds have too many of. No coincidence that the Reds have been better since CPatt was dropped from getting much PT and with Ross seeing more ABs lately. In fact, with Harang pitching poorly, Arroyo and Cueto inconsistently, and the fifth starter search still in progress, as well as injuries to Hairston and Keppinger (and Kepp's mini slump since returning to action)it's a sign of how solid a core this team has that they've managed to play above .500 for several weeks now. The really bad stretch that started on the west coast and finally ended at home before they went 5-4 on the AL trip has killed their record to date. The pitching, hitting and defense were all horrible in that span.

Ltlabner
07-19-2008, 09:02 AM
I hope I'm not the only one that has no clue about what those stats are or mean or anything for that matter.

Sorry Degen, I should have explained them when I first started the thread. I'll post the defination of each stat first and make my comment in itallics. Definitions provided by Baseball Prospectus.

EQA: Equivalent Average. A measure of total offensive value per out, with corrections for league offensive level, home park, and team pitching. EQA considers batting as well as baserunning, but not the value of a position player's defense. The EqA adjusted for all-time also has a correction for league difficulty. The scale is deliberately set to approximate that of batting average. League average EqA is always equal to .260

IMO, I like to consider EQA to be "batting average improved". Basically, higher numbers means the guys produces more runs for every time he makes an out than guys with lower EQA. It's adjusted for time and league and includes baserunning (although I'm not sure specifally what that means). The stat is designed to "look" like batting average in that .260 would be conisdered average, .280ish pretty good, .300 smokin, etc.

VORP: Value Over Replacement Player. The number of runs contributed beyond what a replacement-level player at the same position would contribute if given the same percentage of team plate appearances. VORP scores do not consider the quality of a player's defense.

IOW how many more runs above a scrub does that player produce when compared to other players at the same position. The higher the number the better.

RAR: Runs Above Replacement. RAR compares a hitter's Equivalent Run total to that of a replacement level player who makes the same number of outs and plays the same position

One of the other guys who knows better can chime in here, but my jake-legged defintion is that RAR means how many runs above a scrub does this player produce but it's adjusted for outs. VORP is based on PA's where RAR is more tuned to outs.

RAP: Runs Above Position: The number of Equivalent Runs this player produced, above what an average player at the same postion would have produced in the same number of outs.

I really like RAP. What it is saying is how many runs does the player produce beyond what an average player at the same position does. Not scrub level, but league average. It's also adjusted for outs so it isn't squewd by a guy with 10 PA's versus the one with 300. So the higher the number the more productive the guy is over average players at the same position.

Hope this helps. As I said, I'm no stat expert so I hope I haven't screwed anything up too bad.

But I like these numbers because it tells you a little more than BA/OPS by themselves. I also like the way they are adjusted so players abilities aren't squed by a hot streak and low number of PA's. Bascially, the higher the numbers you see in those categories, the more productive a guy is. I have a hard time remembering much of the other stats so I like to keep it simple. Give me a guy with high numbers in all those categories and I know we have a dude who is getting the job done at the plate.

jojo
07-19-2008, 05:15 PM
Sorry Degen, I should have explained them when I first started the thread. I'll post the defination of each stat first and make my comment in itallics. Definitions provided by Baseball Prospectus.

EQA: Equivalent Average. A measure of total offensive value per out, with corrections for league offensive level, home park, and team pitching. EQA considers batting as well as baserunning, but not the value of a position player's defense. The EqA adjusted for all-time also has a correction for league difficulty. The scale is deliberately set to approximate that of batting average. League average EqA is always equal to .260

IMO, I like to consider EQA to be "batting average improved". Basically, higher numbers means the guys produces more runs for every time he makes an out than guys with lower EQA. It's adjusted for time and league and includes baserunning (although I'm not sure specifally what that means). The stat is designed to "look" like batting average in that .260 would be conisdered average, .280ish pretty good, .300 smokin, etc.

VORP: Value Over Replacement Player. The number of runs contributed beyond what a replacement-level player at the same position would contribute if given the same percentage of team plate appearances. VORP scores do not consider the quality of a player's defense.

IOW how many more runs above a scrub does that player produce when compared to other players at the same position. The higher the number the better.

RAR: Runs Above Replacement. RAR compares a hitter's Equivalent Run total to that of a replacement level player who makes the same number of outs and plays the same position

One of the other guys who knows better can chime in here, but my jake-legged defintion is that RAR means how many runs above a scrub does this player produce but it's adjusted for outs. VORP is based on PA's where RAR is more tuned to outs.

RAP: Runs Above Position: The number of Equivalent Runs this player produced, above what an average player at the same postion would have produced in the same number of outs.

I really like RAP. What it is saying is how many runs does the player produce beyond what an average player at the same position does. Not scrub level, but league average. It's also adjusted for outs so it isn't squewd by a guy with 10 PA's versus the one with 300. So the higher the number the more productive the guy is over average players at the same position.

Hope this helps. As I said, I'm no stat expert so I hope I haven't screwed anything up too bad.

But I like these numbers because it tells you a little more than BA/OPS by themselves. I also like the way they are adjusted so players abilities aren't squed by a hot streak and low number of PA's. Bascially, the higher the numbers you see in those categories, the more productive a guy is. I have a hard time remembering much of the other stats so I like to keep it simple. Give me a guy with high numbers in all those categories and I know we have a dude who is getting the job done at the plate.

I like the notion of looking at a player's worth using both replacement level and average as baselines.

Spring~Fields
07-19-2008, 08:43 PM
I like stats, but I think it tends to get out of hand on occasion as well.

I agree, I also think that it would help if we each spoke the same language in a way.

What good is it to be the "king" of stats if your followers don't have a clue about what your saying or communicating?